MRI-Based Synthetic CT Shows Promise as a Radiation-Free Alternative to Conventional CT in Orthopaedics

Area of Science:

• Medical Imaging
• Radiology
• Biomedical Engineering

Background:

• Computed tomography (CT) is the standard for bone imaging but poses radiation risks, especially for pediatric patients.
• Growing concerns about radiation exposure necessitate the exploration of safer imaging alternatives.

Purpose of the Study:

• To evaluate emerging magnetic resonance imaging (MRI) techniques as radiation-free alternatives to CT for bone imaging.
• To assess the clinical utility of these MRI techniques in pediatric imaging, surgical planning, and cost reduction.

Main Methods:

• Utilized advanced MRI sequences including zero echo time (ZET), ultrashort echo time (USET), and 3-dimensional (3D) gradient recalled echo (GRE).
• Investigated deep learning-based synthetic CT generation as a novel imaging approach.
• Compared the diagnostic accuracy of these MRI modalities against conventional CT across various anatomical locations.

Main Results:

• All evaluated MRI techniques demonstrated high agreement with CT for bone imaging.
• Zero echo time (ZET) MRI provided high-resolution and silent imaging capabilities.
• 3D gradient recalled echo (GRE) MRI is widely available and requires minimal implementation setup.
• Deep learning-based synthetic CT showed potential for generating accurate CT-like images, with capacity for improvement over time.

Conclusions:

• Emerging MRI techniques are viable, radiation-free alternatives to CT for bone imaging.
• These MRI modalities show significant promise for pediatric applications, surgical planning, and potentially reducing healthcare costs.
• Deep learning-based synthetic CT holds substantial future potential, contingent on further data development and validation.